Update to PrePilot Lite

{
  "description": "Use when the user needs ad angles, hooks, creative briefs, test matrices, fatigue diagnosis, message tests, visual concepts, UGC directions, or a structured way to learn which creative idea drives better response.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 235
    }
  ],
  "name": "creative-testing",
  "skill_md_contents": "---\nname: creative-testing\ndescription: Use when the user needs ad angles, hooks, creative briefs, test matrices, fatigue diagnosis, message tests, visual concepts, UGC directions, or a structured way to learn which creative idea drives better response.\n---\n\n# Creative Testing\n\n## Use it for\n\n- Paid creative briefs\n- Hook and angle generation\n- Creative test design\n- Fatigue or performance diagnosis\n- Turning customer insights into ad concepts\n\n## Operating rules\n\n- Use context already present in the conversation before asking for more input.\n- Separate known facts, reasonable assumptions, and unknowns. Do not present assumptions as evidence.\n- When the answer depends on current platform rules, market conditions, pricing, benchmarks, or competitors, verify them with current sources when tools are available.\n- Prefer concrete decisions, examples, and next actions over generic marketing advice.\n- Do not invent campaign performance, customer quotes, research findings, testimonials, rankings, or competitor claims.\n- Keep the requested market, language, funnel stage, audience awareness, budget, and channel constraints visible throughout the work.\n- Test meaningful variables, not cosmetic changes labeled as experiments.\n- Every concept must connect audience tension, promise, proof, and format.\n- Do not invent creator testimonials or product demonstrations.\n- Keep production feasibility and asset rights in scope.\n\n## Workflow\n\n1. Define audience, buying state, offer, platform, current winner/loser signals, and creative constraints.\n2. Generate distinct hypotheses across problem, outcome, proof, mechanism, objection, comparison, or demonstration.\n3. Choose formats that can express each hypothesis clearly.\n4. Build a test matrix with one primary learning goal per cell.\n5. Specify production brief, landing-page message match, metric, minimum read period or evidence threshold, and next action.\n\n## Output contract\n\n- Creative hypothesis bank\n- Prioritized test matrix\n- Production-ready briefs\n- Readback and iteration plan\n\n## Quality gate\n\n- Concepts are meaningfully different\n- Tests answer a decision\n- Proof is real or explicitly needed\n- Metric matches the creative job\n\n## Handoff\n\nIf the task is part of a larger marketing request, return the completed deliverable plus the evidence or decisions the next PrePilot skill needs. Do not repeat upstream analysis unless it changes the result.\n"
}